development of stand-alone green hybrid system for ... - mdpi
TRANSCRIPT
sustainability
Article
Development of Stand-Alone Green Hybrid Systemfor Rural Areas
Solomon Kiros 1, Baseem Khan 2 , Sanjeevikumar Padmanaban 3,* , Hassan Haes Alhelou 4 ,Zbigniew Leonowicz 5 , Om Prakash Mahela 6 and Jens Bo Holm-Nielsen 3
1 Department of Electrical and Computer Engineering, Mekelle University, Mekelle 0231, Ethiopia;[email protected]
2 Department of Electrical and Computer Engineering, Hawassa University, Hawassa 05, Ethiopia;[email protected]
3 Center for Bio-Energy and Green Engineering, Department of Energy Technology, Aalborg University,Niels Bohrs Vej 8, 6700 Esbjerg, Denmark; [email protected]
4 Department of Electrical Power Engineering, Faculty of Mechanical and Electrical Engineering,Tishreen University, Lattakia 2230, Syria; [email protected]
5 Faculty of Electrical Engineering, Wroclaw University of Science and Technology, Wyb. Wyspianskiego 27,50370 Wroclaw, Poland; [email protected]
6 Power System Planning Division, Rajasthan Rajya Vidyut Prasaran Nigam Ltd., Jaipur 302005, India;[email protected]
* Correspondence: [email protected]
Received: 7 April 2020; Accepted: 28 April 2020; Published: 7 May 2020�����������������
Abstract: Despite the tremendous efforts exhibited by various utilities around the world duringthe past few years, there are still exceedingly many remote regions unreached by the electrical grid.For those regions, the enormous available potential of renewable energy resources is believed to beuseful for the development of a stand-alone power supply system. This paper presents the modelingof a stand-alone hybrid system for the remote area of Ethiopia. A comparison of the economicperformance of various scenarios of a stand-alone photovoltaic (PV)-wind hybrid system, with batterystorage and diesel as a backup for electrifying remote rural areas, is presented. Therefore, a practicalexample, Kutur village of Awlio kebele of the Axum district, Ethiopia (which is 30 km away from theclosest national grid) is considered for this research. Two electric load scenarios are estimated byconsidering the set of incandescent and efficient lamps for lighting for the 120 existing households.The above-mentioned solar radiation and wind speed are then used as an input to simulate the hybridset-up for the high and low load estimation using HOMER software. The simulation result showsthat the net present costs (NPC) corresponding to the high and low load scenarios is $262,470 and$180,731, respectively. Besides, an essential load forecasting is performed to see the effect of theincrease in electric demand of the community on the required investment to install a stand-alonehybrid set-up. The NPC after load forecasting is found to be more than three folds of the NPC requiredfor the reference year. In both cases, the simulation results indicate that using a stand-alone PV-windhybrid system with battery storage and a diesel generator as a backup for electrifying Kutur villageis cost-effective and comparable against the cost required for electrifying the village by extendingthe grid.
Keywords: grid access; hybrid PV-wind; battery storage; diesel generator; HOMER
1. Introduction
Renewable energy generation sources are essential for providing green energy to remote ruralareas. Development of green energy-based stand-alone systems is possible with the proper utilization
Sustainability 2020, 12, 3808; doi:10.3390/su12093808 www.mdpi.com/journal/sustainability
Sustainability 2020, 12, 3808 2 of 14
of solar photovoltaic (PV) and wind energy sources. The electric supply system throughout the worldis the interconnected system (ICS). Therefore, most of the remote rural areas that are located far off fromthe grid did not get electric access since the power to the grid is insufficient and it is not cost-effective toextend the medium voltage distribution system, even though the development cost within the countryis cheap. Consequently, the rural areas have been dependent on local solutions for electricity supply.These areas have been using traditional biomass as a source of energy for baking as well as cookingand oil for lighting purpose. However, the current price increase in imported oil and the harmfuleffects of fossil fuels on the local and global environment motivate the search for other alternatives.
The Ethiopian interconnected system (ICS) consists of fourteen hydro, six diesel standbys, onegeothermal, three wind farms and one biomass/waste power plants with the installed capacity of3814.20 MW, 99.17 MW, 7.30 MW, 324 MW and 25 MW, respectively, which bring a total of 4269.67 MW [1].Despite the abundance of potential resources suitable for the energy sector development, the level ofelectricity production can improve from the current capacity of 4269.67 MW. As a result, the overallaccess to electricity is increased to 35% from 13% [1]. The government also has a plan to exploit moreenergy from wind and geothermal sources to improve access to electricity by up to 50% in the comingfive years. Ethiopian Electric Power (EEP), the sole electric power producer and supplier in the country,is therefore endowed to fulfil the plan by increasing the production up to 10,000 MW by includingrenewable energy resources and especially wind [2]. Moreover, there is also a program to give electricaccess from solar resources to more than 150,000 households of remote rural areas of the country withsufficient solar radiation resource in the coming five years [2].
Despite the availability of the generation, the rural areas of Ethiopia are facing the problem of noelectricity access due to various financial and demographic limitations in the grid extension. However,these limitations can be minimized by looking for alternative energy resources, which can be used asstand-alone systems for providing electric access to remote communities. Therefore, solar and windcan be the first option, since these resources are available in those areas free of cost.
Related Work
Various studies performed on the hybrid systems in different rural areas of the world showed thatthe hybrid set-up is cost-effective and reasonable for electrifying rural areas. This was demonstratedusing different approaches, such as analytical models including life cycle cost (LCC) and loss ofpower supply probability (LPSP). Additionally, researchers used various software’s like MATLAB andHOMER to obtain the best-optimized result for the specific area in which the research work has beenconducted. Salwan [3] proposed a hybrid system as a renewable resource of power generation forgrid-connected applications in three cities in Iraq. The proposed system was simulated using MATLABsolver. Barley et al. [4] used hourly values of wind speed and solar radiation for one year for theirmodels to capture seasonal effects, diurnal cycles, storm cycles and stochastic variations. For thatpurpose, a quasi-steady-state time-series model was utilized to determine approximate least-costdesigns and a stochastic model for accurately determining the cost of energy (COE) and unmet loadfor each of the designs indicated by the simple model. Finally, actual resource data from the regionwere processed to indicate that the combination of wind and PV is more cost-effective than eithercomponent alone. Fieldler et al. [5] studied PV-wind hybrid systems for eleven locations in Sweden.They aimed for the evaluation of system cost, cost of energy, the effect of load size, and compared thecost of the hybrid system against the cost required by a PV system alone. The system was modeled tosupply electricity for single-family houses. HOMER software was used for the sizing of the hybridsystems based on the net present costs (NPC). Mandal et al. [6] presented the model for optimalsizing of the stand-alone hybrid system to electrify remote areas of Bangladesh. Aberilla et al. [7]designed and assessed the environmental sustainability of a low-level stand-alone energy systemto electrify rural communities in the Philippines. Zhao et al. [8] assessed the technical feasibility ofoff-grid PV-wind/adiabatic compressed air energy storage depended on the hybrid power systemto electrify a remote mobile base station. Moner-Girona et al. [9] presented a case study related to
Sustainability 2020, 12, 3808 3 of 14
the decentralization of the rural electrification in Kenya for speeding up the universal energy accessprogram. Gabra et al. [10] compared the techno-economic analysis of off-grid wind microgrids with PVand diesel-based systems in Africa. Ariyo et al. [11] optimized the performance of an off-grid hybridpower system developed for the senate building, university of Ilorin, Nigeria. Yeshalem et al. [12]designed a stand-alone hybrid PV-wind power system for a remote mobile base station in Ethiopia.Jariso et al. [13] developed an off-grid PV energy system to supply electricity to a remote health centerin south-west Ethiopia. Further, in [14], Jariso et al. optimized the performance of a hybrid systemdeveloped for a health center south-west Ethiopia. After the critical review of the mentioned works ofliterature, it is concluded that many of them have not utilized load forecasting to show the variationof the demand on the installation cost. On top of that, many did not compare the total cost of theextension of national grid; instead, compared the hybrid system with one of the components alone.
Henceforth, this research is conducted to show the impact of population growth on the demandof the site by using load forecasting as well as to compare the cost of the hybrid system againstthe cost required to extend the national grid. The following are the significant contributions of thisresearch paper:
1. Modeling of a stand-alone hybrid PV-wind hybrid system, with battery storage and diesel as abackup for electrifying remote rural areas of Ethiopia;
2. Actual local metrological solar irradiation and wind speed data are considered to assess theenergy outputs from the distributed generation sources;
3. A comparison of the economic performance of various scenarios of the stand-alone PV-windhybrid system, with battery storage and diesel as a backup for electrifying remote rural areasis presented;
4. Basic load forecasting is performed to see the effect of the increase in electric demand of thecommunity on the required investment to install a stand-alone hybrid set-up;
5. A comparative analysis of the cost of the hybrid system against the cost required to extend thenational grid is performed so as to provide the best economic signals.
The rest of the manuscript is organized as follows. Section 2 provides detailed information aboutthe selected case study area. Section 3 discusses the methodology utilized to perform the researchwork. Section 4 presents the resource assessment and load estimation of the selected case study area.Section 5 discusses the modeling of the hybrid system and cost analysis of the grid extension. Section 6presents the result and discussion followed by the conclusion.
2. Study Area
Kutur village of Awlio kebele (ward) of the Axum district is located 30 km away from the closestnational grid. The village was selected due to the possibility of its grid extension, and hence thecomparison of the installation cost of a stand-alone green system with an extended grid system couldbe performed. Axum district is located at the latitude of 14◦07’N and longitude of 38◦43’E and found tohave 4.28 m/s wind speed at the height of 10 m, and 6.19 kWh/m2/day solar radiation. Two electric loadscenarios were estimated by considering the set of incandescent and efficient lamps for the lighting of120 existing households.
Figure 1 shows the settlement of the people of Kutur village of Awlio Kebele found in Axumdistrict and their traditional use of firewood for baking. As is shown in the figure, the area has asewing workshop. In addition to this, it has small shops, which supply the daily requirements ofthe community. It also shows the traditional “mitad” or stove used for baking bread and “Injera”.Firewood is the primary source of energy, and hence the forests found in the area are highly exposed todeforestation. Therefore, this work also shows the possibility of using a PV-wind hybrid system forelectrifying the village and its significance in reducing deforestation caused by the people of the area.
Sustainability 2020, 12, 3808 4 of 14
Sustainability 2020, 12, x FOR PEER REVIEW 4 of 1
Sustainability 2020, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/sustainabilit
Figure 1. Kutur Village.
3. Methodology
As was mentioned in the introduction, this research is mainly concerned with looking into the better options of energy sources to electrify villages in remote areas of Ethiopia that are off the existing national grid by comparing against the cost needed to extend the national grid. The main concern is the assessment of resource potential and the modeling of the stand-alone hybrid system by analyzing data taken from different authorities. The field data, which include the number of households and their daily activity related with energy consumption, were collected by visiting the selected village, and some pictures were taken, which show the settlement and the wood the residents use as the source of energy. However, wind speed and solar radiation data of the selected sites collected from National Aeronautics and Space Administration (NASA) were compared with data from the National Meteorology Service Agency (NMSA) of Ethiopia due to their better accuracy and understanding of the demographic conditions and incorporation of the local weather effects, if any. Moreover, the data needed for calculating the cost of extending the grid were taken from EEP, and were analyzed using an interlinked Microsoft office excel spreadsheet prepared by the universal energy access program (UEAP). Additionally, the exponential load forecasting method was used to consider the effect of change of electric load of the community within the lifespan of the project, which is 25 years. Finally, the hybrid system was modeled and simulated using HOMER software, based on the primary and secondary analyzed data so as to get the best optimization. Figure 2 presents a flow chart of the research progress.
Figure 1. Kutur Village.
3. Methodology
As was mentioned in the introduction, this research is mainly concerned with looking into thebetter options of energy sources to electrify villages in remote areas of Ethiopia that are off the existingnational grid by comparing against the cost needed to extend the national grid. The main concern isthe assessment of resource potential and the modeling of the stand-alone hybrid system by analyzingdata taken from different authorities. The field data, which include the number of households andtheir daily activity related with energy consumption, were collected by visiting the selected village,and some pictures were taken, which show the settlement and the wood the residents use as thesource of energy. However, wind speed and solar radiation data of the selected sites collected fromNational Aeronautics and Space Administration (NASA) were compared with data from the NationalMeteorology Service Agency (NMSA) of Ethiopia due to their better accuracy and understanding ofthe demographic conditions and incorporation of the local weather effects, if any. Moreover, the dataneeded for calculating the cost of extending the grid were taken from EEP, and were analyzed using aninterlinked Microsoft office excel spreadsheet prepared by the universal energy access program (UEAP).Additionally, the exponential load forecasting method was used to consider the effect of change ofelectric load of the community within the lifespan of the project, which is 25 years. Finally, the hybridsystem was modeled and simulated using HOMER software, based on the primary and secondaryanalyzed data so as to get the best optimization. Figure 2 presents a flow chart of the research progress.
Sustainability 2020, 12, 3808 5 of 14
Sustainability 2020, 12, x FOR PEER REVIEW 5 of 1
Sustainability 2020, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/sustainabilit
Figure 2. Flow chart of the research process.
4. Resource assessment and Load Estimation
The solar data available from most of the towns of Ethiopia relate to sunshine duration, and the wind data are related to the wind speed measured at the height of 2 m. In order to assess the availability of solar radiation and wind speed resources in Axum district, data were taken from different sources, including the NMSA. The data obtained from NMSA were then compared against the data from NASA, to check for its accuracy and a better understanding of the demographic conditions. Further, local weather effects, if any, were incorporated.
4.1. Assessment of Solar Radiation
The assessment of the potential for the solar radiation of the selected area was done by taking data from different sources. These are NASA and NMSA of Ethiopia, as well as data assessed by [2] for the regions of the country. In the first approach, the data taken from the NMSA were analyzed using a mathematical approach based on the sunshine duration collected in the past six consecutive years. The sunshine duration data taken from the NMSA of Ethiopia, Mekelle branch, were converted into solar radiation using the Angstrom radiation–sunshine relation presented in [15].
The measurements were taken for six consecutive years. As can be seen in Table 1, the data for years 2010, 2012 and 2013 are full, but for 2009, 2011 and 2014 measurements were taken partially; hence, the three years with full data were taken to convert into solar radiation. Therefore, for calculating the radiation in kWh/m2 from the sunshine duration using the Angstrom radiation–sunshine relation [15], the values for the regression coefficients α and β were taken to be
Figure 2. Flow chart of the research process.
4. Resource Assessment and Load Estimation
The solar data available from most of the towns of Ethiopia relate to sunshine duration, and thewind data are related to the wind speed measured at the height of 2 m. In order to assess the availabilityof solar radiation and wind speed resources in Axum district, data were taken from different sources,including the NMSA. The data obtained from NMSA were then compared against the data from NASA,to check for its accuracy and a better understanding of the demographic conditions. Further, localweather effects, if any, were incorporated.
4.1. Assessment of Solar Radiation
The assessment of the potential for the solar radiation of the selected area was done by takingdata from different sources. These are NASA and NMSA of Ethiopia, as well as data assessed by [2] forthe regions of the country. In the first approach, the data taken from the NMSA were analyzed using amathematical approach based on the sunshine duration collected in the past six consecutive years.The sunshine duration data taken from the NMSA of Ethiopia, Mekelle branch, were converted intosolar radiation using the Angstrom radiation–sunshine relation presented in [15].
The measurements were taken for six consecutive years. As can be seen in Table 1, the data foryears 2010, 2012 and 2013 are full, but for 2009, 2011 and 2014 measurements were taken partially; hence,the three years with full data were taken to convert into solar radiation. Therefore, for calculating theradiation in kWh/m2 from the sunshine duration using the Angstrom radiation–sunshine relation [15],the values for the regression coefficients α and β were taken to be 0.30 and 0.50, respectively. Finally,
Sustainability 2020, 12, 3808 6 of 14
this result was compared with the data taken from NASA, and it was found that the annual averagesolar radiation obtained from NMSA (which is 6.21 kWh/m2) is nearly equal to the value taken fromNASA (which is 6.15 kWh/m2). Consequently, 6.21 kWh/m2 was then taken as an input to the solarresource in the modeling of the hybrid system. The monthly average solar radiation of Axum for thedata taken from NMSA and NASA is shown in Tables 1 and 2, respectively.
Table 1. Solar radiation of Axum in kWh/m2/day data from NMSA.
Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec2010 6.11 6.39 7.22 6.94 7.22 6.11 5.28 5.56 6.39 6.67 6.11 5.832012 6.11 6.11 6.94 7.5 7.5 6.11 4.72 5.56 6.67 6.67 5.83 5.832013 5.83 6.39 6.67 6.67 6.67 5.83 5.28 4.44 5.83 6.67 6.11 5.83
Annual Averagein kWh/m2/day
Monthly average 6.02 6.3 6.94 7.04 7.13 6.02 5.09 5.19 6.3 6.67 6.02 5.83 6.21
Table 2. Solar radiation of Axum in kWh/m2/day data from NASA.
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Annual Average22-year Average
5.76 6.2 6.59 6.92 6.74 6.49 5.8 5.58 6.25 6.23 5.88 5.51 6.15
From the results shown in Table 1 it can be seen that the solar radiation of the area is highest in May,reaching a value of 7.13 kWh/m2/day, and is lowest in July (which has a value of 5.09 kWh/m2/day).Hence, depending on the results obtained from NMSA and NASA, it can be concluded that the areahas excellent potential for using the photovoltaic system for rural electrification.
Figure 3 presents the monthly distribution of solar radiation of Axum city according to the dataobtained from NMSA.
Sustainability 2020, 12, x FOR PEER REVIEW 6 of 1
Sustainability 2020, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/sustainabilit
0.30 and 0.50, respectively. Finally, this result was compared with the data taken from NASA, and it was found that the annual average solar radiation obtained from NMSA (which is 6.21 kWh/m2) is nearly equal to the value taken from NASA (which is 6.15 kWh/m2). Consequently, 6.21 kWh/m2 was then taken as an input to the solar resource in the modeling of the hybrid system. The monthly average solar radiation of Axum for the data taken from NMSA and NASA is shown in Table 1 and Table 2, respectively.
Table 1. Solar radiation of Axum in kWh/m2/day data from NMSA.
Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Annual Average in
kWh/m2/day 2010 6.11 6.39 7.22 6.94 7.22 6.11 5.28 5.56 6.39 6.67 6.11 5.83 2012 6.11 6.11 6.94 7.5 7.5 6.11 4.72 5.56 6.67 6.67 5.83 5.83 2013 5.83 6.39 6.67 6.67 6.67 5.83 5.28 4.44 5.83 6.67 6.11 5.83
Monthly average
6.02 6.3 6.94 7.04 7.13 6.02 5.09 5.19 6.3 6.67 6.02 5.83 6.21
From the results shown in Table 1 it can be seen that the solar radiation of the area is highest in May, reaching a value of 7.13 kWh/m2/day, and is lowest in July (which has a value of 5.09 kWh/m2/day). Hence, depending on the results obtained from NMSA and NASA, it can be concluded that the area has excellent potential for using the photovoltaic system for rural electrification.
Table 2. Solar radiation of Axum in kWh/m2/day data from NASA.
22-year Average Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Annual Average 5.76 6.2 6.59 6.92 6.74 6.49 5.8 5.58 6.25 6.23 5.88 5.51 6.15
Figure 3 presents the monthly distribution of solar radiation of Axum city according to the data obtained from NMSA.
Figure 3. Solar radiation of Axum (data from NMSA).
4.2. Assessment of Wind Potential
To maintain the current economic growth of Ethiopia, it needs a 38-fold increase in its electric supply by 2030; hence, new sources of energy are urgently needed. Subsequently, the answer to this may lie in wind energy, since intense and reliable winds of the country can generate a substantial amount of electricity at a reasonable cost, which can even be exported to neighboring countries. The areas that can provide the highest potential for wind energy are the northern, central, eastern and southwestern part of the country [16]. Axum is also one of the districts found in the northern part of Ethiopia, which has good potential in wind and solar resources. However, to assess the potential of wind energy of the area, the data were simply taken from NASA, because of the absence of full data from the NMSA. Therefore, the annual average wind speed for Axum was found to be 4.28 m/s at a height of 10 m. This value was used
Figure 3. Solar radiation of Axum (data from NMSA).
4.2. Assessment of Wind Potential
To maintain the current economic growth of Ethiopia, it needs a 38-fold increase in its electricsupply by 2030; hence, new sources of energy are urgently needed. Subsequently, the answer to thismay lie in wind energy, since intense and reliable winds of the country can generate a substantialamount of electricity at a reasonable cost, which can even be exported to neighboring countries.The areas that can provide the highest potential for wind energy are the northern, central, eastern andsouthwestern part of the country [16]. Axum is also one of the districts found in the northern part ofEthiopia, which has good potential in wind and solar resources. However, to assess the potential ofwind energy of the area, the data were simply taken from NASA, because of the absence of full datafrom the NMSA. Therefore, the annual average wind speed for Axum was found to be 4.28 m/s at aheight of 10 m. This value was used as the input to the wind resource in the modeling of the hybridsystem. Table 3 presents the 10 years’ monthly average wind speed of Axum city at the height of 10 m.
Sustainability 2020, 12, 3808 7 of 14
Table 3. Monthly average wind speed of Axum at the height of 10 m.
Lat 14 Long 38.7 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Annual Average10-year average 4.45 4.48 4.46 4.26 3.64 4.92 5.42 5 4.08 3.13 3.56 4.01 4.28
4.3. Load Estimation
The load estimation, which is the essential step in modeling the hybrid system, was done for asingle village found in Axum district. The name of the area is Awlio, and it is around 30 km away fromthe existing grid found nearby, and 55 km from Adwa substation. Awlio kebele has 942 householdfamilies with a total population of 4274. One of the villages of this kebele (Kutur) has one primaryschool, the kebele administration office, Telecom office, a farmers training center (FTC) and one healthcenter. The health center (HC) has midwives to help mothers deliver babies.
Moreover, HC offers support and advice to pregnant women. The load estimation for HC, FTC andkebele administration was done based on the Universal Electric Access Program (UEAP) standard [17].The population of this village accounts for around 13% of the total population of the kebele with about120 households.
The deferrable load contains one water pump for all households as well as the service centers thatdeliver public service to the community. The type of water pump is HR-14 taken from the Lorenz PS600series, with a 300 W power rating and a pumping capacity of 40 L/m (i.e., 2400 L/h) [18]. The averagedaily water consumption of the community per household is 140 L/day, including the water usedfor their cattle. The total water consumption of the community is then 16,800 L/day. Additionally,all the service centers are assumed to use a summed average of 4000 L/day. This indicates that theoverall water needed by the village is 20,800 L/day, and four water pumps of the types mentionedabove are used to supply the water demand of the community. The water storage tank is assumed tohave a capacity of storing water for four days (i.e., 83,200 L of water). The number of hours needed tosupply the daily water consumption of the community was obtained to be approximately 2.5 h. Hence,the daily deferrable load is 3 kWh/day, and the corresponding electricity storage capacity is 10.8 kWh.
The primary load of Kutur village was calculated by considering the following: the lamps of 60 Wfor lighting systems with three lamps for each household; a radio receiver of 20 W; a flour mill with arating of 15 kW; and a TV and DVD with ratings of 65 W and 20 W, respectively; but the TV and DVDare assumed to be used by the model farmers of the community. Their number is assumed to be 10% ofthe total households of the village. The detailed initial load estimation is as shown in Table 4. On theother hand, the load estimations for HC, FTC and kebele administration as stated earlier were takenfrom UEAP (which are 3 kW each) [18], because it is difficult to estimate their load based on the presentcondition. The total power demand for the three service centers was thus taken to be 9 kW. The dailypower demand of the community, as shown in Table 4, is 51.10 kW, and the energy is 307.68 kWhwithout the deferrable load.
Table 4. Primary load estimation of Kutur village.
Primary Load SingleHousehold
No. ofHouseholds
Total No. ofEquipment
Size in W for1 Equipment
Total Powerin kW
Operating hrsper Day
Energyin kWh Remark
Lighting lamps 3 120 360 60 21.6 6 129.6Radio (tape recorder) 1 120 120 20 2.4 8 19.2
TV 1 12 12 65 0.78 6 4.68DVD 1 12 12 20 0.24 4 0.96Mill 1 1 15,000 15 5 75
Health center (HC) 3 8 24Farmers training center
(FTC) 3 8 24
Kebele Adm. 3 8 24Primary school 4 13 52 40 2.08 3 6.24 FL
Total 51.1 307.68
Sustainability 2020, 12, 3808 8 of 14
5. Modeling of the Hybrid System and Cost Analysis of Grid Extension
5.1. Modelling of the Hybrid System
As is shown in the assessment of solar radiation, Axum has a monthly average solar radiationvarying between a minimum value of 5.09 kWh/m2 and a maximum value of 7.13 kWh/m2, with anannual average of 6.21 kWh/m2. Similarly, the monthly average wind speed of Axum varies between3.13 m/s to 5.42 m/s at the height of 10 m, with an annual average of 4.28 m/s. These numbers indicatethat the area has a potential for implementing a PV-wind hybrid system to give electric access to thecommunity of Kutur village of Awlio kebele found in Axum district. However, the investment costs ofPV and wind turbines have always been the main barrier to the use of the hybrid system for small-scaleas well as large-scale applications. However, due to technology advancement, the costs of PV systemsand wind turbines are decreasing, while the price of oil is increasing in addition to the depletion of oilresources. This is encouraging developing countries like Ethiopia, which have excellent resources inboth systems, to use a stand-alone hybrid system for supplying electric access to remote rural areas.The current total investment cost of PV has reached 2625 $/kW, and for that of wind, it has decreased to1100 $/kW [19]. The current electricity grid coverage of Ethiopia is around 40%. Hence, the hybridsystem can be competitive irrespective of its initial capital cost when considering the rapid increase inoil prices. Moreover, it has a negligible impact on the global and local environment. Further, solar aswell as wind power are becoming serious candidates in the electricity market due to increasing oilprices and the substantial increase in manufacturing capacity of wind turbines and solar modules [20].According to a green energy report conducted in Denmark, an estimated 23.9 GW of cells and 20 GWof modules were produced in 2010 by the solar PV industry [20]. Moreover, the total global capacity ofwind power and solar photovoltaic power has reached 198 GW and 40 GW, respectively [20]. However,the PV-wind hybrid system may not be sufficient to supply energy for 24 h for the whole year, andtherefore has to be supported by diesel generators and batteries, which can be used as a backupfor supplying sustainable electricity when using a PV-wind hybrid system. The cost of generatorsvaries between 200 $/kW to 1000 $/kW [21]. Therefore, in this work, the resource potential of Axum isassessed, and a PV-wind hybrid system with diesel and battery as a backup to electrify 120 householdsof Kutur village is modeled. Additionally, the initial capital cost of the hybrid system is comparedagainst the cost required for extending the grid.
5.2. Cost Analysis of Grid Extension
Power is transmitted to the end-users through distribution lines. The mission of the UniversalElectric Access Program (UEAP) is to deliver power to consumers in the remote rural areas of thecountry. It is typically done by extending the national grid from a nearby substation or an overheadmedium voltage (MV) distribution line. The medium voltage levels that are most commonly usedin MV line construction by the UEAP are 33 kV, 19 kV and 15 kV. Currently, the UEAP is using the33 kV voltage level for electrifying the remote rural areas due to its advantage over the other options.The main components of an overhead distribution line that can affect the cost of grid extension are theline supports and their foundation, insulators, line accessories, conductors and overhead earth wires.The line supports can be made from wood (like local wood poles and South Africa poles), reinforcedconcrete poles and steel, but due to its strength and availability, UEAP is using concrete poles for MVdistribution lines to give electric access for selected rural areas.
The Universal Electric Access program is currently electrifying the rural areas by extending thegrid found nearby them. The main tasks done during extension are surveying, excavation, poleerection, pole top configuration and stringing of conductors. For the MV distribution line construction,all aluminium alloy conductor with a size of 95 mm2 is used for the 33 kV line. However, the number ofpoles to be erected and the wire required to depend on the distance of the site from a nearby substationor MV distribution line, as well as the geographical location of the selected sites. Moreover, differentaccessories are used depending on the type of assemblies; these can be classified as suspension, light
Sustainability 2020, 12, 3808 9 of 14
angle 1, light angle 2, heavy angle, T-off, tension tower and dead-end assemblies based on the angledeviation from a straight line.
Based on the theory mentioned earlier, the cost estimation for delivering electric access to Kuturvillage, which is found in Axum district, was performed. The distance of Kutur from an existing grid(i.e., Axum) is 30 km, and a rough survey was conducted to select the type of assemblies to be used forMV distribution line construction. This village needs 1 × 200 kVA rating transformers for supplyingpower to the community.
To give electric access to the 120 households, HC, FTC, flour mill, kebele administration andprimary school of Kutur from the national grid found nearby, an extension of 30 km MV, 5.35 kmlow voltage (LV) distribution lines and erection of one transformer with a rating of 200 kVA areneeded. Hence, the cost is analyzed using an interlinked spread Microsoft excel sheet used by theUEAP; the result obtained is as shown in Table 5. The total cost, as indicated in the table, includesmaterial, transportation and labour costs, and the investment cost is therefore 5,801,286.51 Ethiopianbirr, which is equivalent to 337,284.099 US dollars.
Table 5. Total cost, in Ethiopian Birr, needed for extending the grid to Kutur village of Awlio kebelefound in Axum district.
Description Salary and Wage Allowance Transport Material Total33 kV line extension 555,518.52 678,967.08 339,100.20 2,773,182.82 4,346,768.621 × 200 kVA erection 2589.10 3164.46 3371.28 88,721.63 97,846.471 × load break switch 1516.70 1853.74 919.14 17,295.56 21,585.14
LV line extension 84,309.18 103,044.56 95,114.55 525,228.32 807,696.60Total 643,933.50 787,029.84 438,505.17 3,404,428.32 5,273,896.83
Interest to be capitalizedOver head cost (10%) 527,389.68
Total cost 5,801,286.51Rechargeable/recoverable amount (*contributions)
Net total 5,801,286.51
6. Results and Discussion
HOMER software was used for simulating the hybrid PV-wind system with diesel and batterystorage as a backup. To get the optimal combination of the hybrid components, which could beimplemented as a hybrid system model to give electric access to the community of Kutur village,the model was run for various iterations using different values of the essential variables. The dieselprice was taken to be 1.1 $/L, which is the current price of oil in the country. The interest rate was alsotaken to be 15%, and the project lifetime was set to 25 years. Table 6 presents the various inputs to theHOMER software.
Table 6. Inputs to the HOMER software.
Item Size Capital Replacement Operation andMaintenance Cost
Sizes (kW)Considered
QuantitiesConsidered Lifetime
AC wind turbine(Fuhrlander 30) [2] 30 kW 33,000 24,750 660 0–4 25
AC Generator [21] 12 kW 4000 2560 0.15 0, 12, 24, 36,48, 60 30,000 h
PV [2] 1 kW 2625 2625 0 0–100 25Battery (S4KS25P) [12] 1900 Ah 1250 1100 25 0–100
Converter [12] 1 kW 800 750 8 0–100 15
Results are displayed for the inputs either in the overall form, in which the top-ranked systemconfigurations are listed according to their NPC, or in categorized form, in which only the least costsystem configuration is considered for each possible combination. The variables listed in Table 6 areused as an input for simulating the hybrid set-up using HOMER. As a result, the outputs from thesoftware are obtained in a categorized form. This is shown in Tables 7 and 8, before and after loadforecasting, respectively.
Sustainability 2020, 12, 3808 10 of 14
Table 7. Simulation Results in the categorized form before load forecasting using HOMER.
PV(kW) FL30 GEN
(kW) S4KS25P Converter(kW)
DispatchStrategy
InitialCapital ($)
OperatingCost ($/yr)
TotalNPC ($)
COE($/kWh)
RenewableFraction
CapacityShortage
Diesel(L) GEN (h)
10 3 12 30 25 LF 186,750 11,714 262,470 0.402 0.92 0.09 6020 21123 12 55 25 LF 191,750 12,654 273,549 0.423 0.92 0.09 5619 1941
30 3 40 25 CC 247,750 4828 278,956 0.434 1 0.11 24 LF 41,000 40,811 304,812 0.446 0.41 0.04 31,905 6000
5 1 24 5 LF 58,125 39,029 310,412 0.452 0.46 0.02 30,372 582910 24 5 5 CC 44,500 42,174 317,116 0.461 0.16 0.01 33,627 5147
24 15 5 CC 30,750 47,731 339,289 0.493 0 0.01 37,798 52725 24 5 LF 25,125 48,983 341,758 0.501 0.08 0.06 39,062 6601
24 LF 8000 51,689 342,128 0.512 0 0.08 41,312 693585 80 30 CC 347,125 5273 381,214 0.601 1 0.1
Table 8. Simulation Results in categorized form after load forecasting using HOMER.
PV(kW) FL30 GEN
(kW) S4KS25P Converter(kW)
DispatchStrategy
InitialCapital ($)
OperatingCost ($/yr)
TotalNPC ($)
COE($/kWh)
RenewableFraction
CapacityShortage
Diesel(L)
GEN(h)
20 4 36 45 35 LF 280,750 44,174 566,296 0.38 0.77 0.09 30,525 37264 48 35 30 LF 215,750 60,989 609,989 0.398 0.66 0.05 44,411 41483 48 LF 115,000 81,537 642,069 0.423 0.51 0.08 63,324 5816
10 3 48 10 LF 149,250 78,510 656,753 0.428 0.54 0.07 60,720 563930 48 20 10 LF 127,750 92,636 726,561 0.479 0.21 0.09 73,174 591635 48 20 LF 123,875 95,842 743,414 0.492 0.23 0.1 76,270 6445
60 20 10 CC 53,000 108,470 754,167 0.484 0 0 86,354 511160 LF 20,000 124,440 824,402 0.531 0 0.03 99,223 6935
20 4 36 45 35 LF 280,750 44,174 566,296 0.38 0.77 0.09 30,525 3726
By observing the few set-ups listed on the categorized form, the following results were to befound. The most cost-effective system with the lowest NPC from Table 7 is the PV-wind–generator–battery–converter set-up, where the battery operates using load following (LF), which produces onlyenough power to meet the demand. However, serving the deferrable load is left to the renewableresources cycle charging (CC) strategy. For this set-up, the total NPC is $262,470, and the cost of energy(COE) is 0.402 $/kWh, with a 92% contribution from renewable energy resources. Similarly, the othercost-effective option has 100% renewable resource contribution. However, the battery operates usingthe CC strategy, whereby the battery operates at full output power to serve the primary load, and anysurplus electrical production goes toward the lower priority objectives. For this set-up, the net presentcost is $278,956, and the cost of energy is 0.434 $/kWh.
On the other hand, the results from the categorized form listed in Table 8 were used to look at theother possible optimum combinations after the result of load forecasting. The first set-up containingPV-wind-Generator-Battery-Converter has a renewable resource proportion of 77%. For this set-up,the NPC is $566,296, and COE is 0.38 $/kWh, and this can also be a right choice for the implementation.The second set-up, with a 66% renewable resource contribution, which has an NPC of $609,989 anda COE of 0.398 $/kWh, can be another choice for the implementation. The contribution of monthlyaverage electric production of PV, wind and diesel generator units for the 92% renewable resource isshown in Figure 4. In addition to this, Figure 5 shows the cost summary for the 92% renewable fraction.
Sustainability 2020, 12, x FOR PEER REVIEW 11 of 14
Sustainability 2020, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/sustainability
units for the 92% renewable resource is shown in Figure 4. In addition to this, Figure 5 shows the cost summary for the 92% renewable fraction.
A bar-graph illustrates the cost breakdown for the 92% renewable fraction on the cost summary figure (i.e., Figure 5). Finally, a sensitivity analysis was carried out by varying wind speed, diesel oil price and the PV capital cost multiplier linked with the PV replacement cost multiplier. The wind speed was varied between 2.5 m/s and 4.5 m/s. The price of diesel oil is taken to be 1.1 $/litre, which is the current price within the country. Therefore, the results for this analysis are shown in Figure 6 and Figure 7 for the respective sensitivities of wind speed to the price of diesel and PV capital cost multiplier to diesel price.
Figure 4. Contribution of the power units for 92% utilization of renewable resources.
Figure 5. Cost summary for the 92% renewable resource fraction.
Figure 4. Contribution of the power units for 92% utilization of renewable resources.
Sustainability 2020, 12, 3808 11 of 14
Sustainability 2020, 12, x FOR PEER REVIEW 11 of 14
Sustainability 2020, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/sustainability
units for the 92% renewable resource is shown in Figure 4. In addition to this, Figure 5 shows the cost summary for the 92% renewable fraction.
A bar-graph illustrates the cost breakdown for the 92% renewable fraction on the cost summary figure (i.e., Figure 5). Finally, a sensitivity analysis was carried out by varying wind speed, diesel oil price and the PV capital cost multiplier linked with the PV replacement cost multiplier. The wind speed was varied between 2.5 m/s and 4.5 m/s. The price of diesel oil is taken to be 1.1 $/litre, which is the current price within the country. Therefore, the results for this analysis are shown in Figure 6 and Figure 7 for the respective sensitivities of wind speed to the price of diesel and PV capital cost multiplier to diesel price.
Figure 4. Contribution of the power units for 92% utilization of renewable resources.
Figure 5. Cost summary for the 92% renewable resource fraction. Figure 5. Cost summary for the 92% renewable resource fraction.
A bar-graph illustrates the cost breakdown for the 92% renewable fraction on the cost summaryfigure (i.e., Figure 5). Finally, a sensitivity analysis was carried out by varying wind speed, diesel oilprice and the PV capital cost multiplier linked with the PV replacement cost multiplier. The windspeed was varied between 2.5 m/s and 4.5 m/s. The price of diesel oil is taken to be 1.1 $/litre, which isthe current price within the country. Therefore, the results for this analysis are shown in Figures 6and 7 for the respective sensitivities of wind speed to the price of diesel and PV capital cost multiplierto diesel price.
HOMER software was also used to compare the cost of the hybrid system against the cost requiredfor extending the grid. Consequently, it was found that the hybrid system is cost-effective for an areawhich is found at a distance greater than 10.4 km (which is the breakeven grid extension distance)from the existing grid found nearby.
Sustainability 2020, 12, x FOR PEER REVIEW 12 of 14
Sustainability 2020, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/sustainability
Figure 6. The sensitivity of wind speed to diesel price.
Figure 7. The sensitivity of PV cost to diesel price.
HOMER software was also used to compare the cost of the hybrid system against the cost required for extending the grid. Consequently, it was found that the hybrid system is cost-effective for an area which is found at a distance greater than 10.4 km (which is the breakeven grid extension distance) from the existing grid found nearby.
Figure 6. The sensitivity of wind speed to diesel price.
Sustainability 2020, 12, 3808 12 of 14
Sustainability 2020, 12, x FOR PEER REVIEW 12 of 14
Sustainability 2020, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/sustainability
Figure 6. The sensitivity of wind speed to diesel price.
Figure 7. The sensitivity of PV cost to diesel price.
HOMER software was also used to compare the cost of the hybrid system against the cost required for extending the grid. Consequently, it was found that the hybrid system is cost-effective for an area which is found at a distance greater than 10.4 km (which is the breakeven grid extension distance) from the existing grid found nearby.
Figure 7. The sensitivity of PV cost to diesel price.
7. Conclusions
The hybrid system was simulated based on the input data gathered from the selected area, and wasfound to be cost-effective to use for electrifying Kutur village of Awlio Kebele found in Axum districtby comparing its cost against the cost required to electrify the area by extending the national grid.The results show a number of possibilities with different levels of renewable energy resource utilizationfor the implementation. However, the most cost-effective system is the PV-wind–battery–converterset-up, with a total NPC of $262,470 and COE of 0.402 $/kWh. This system has an excess electricitygeneration of 44.6%, a capacity shortage of 8.9% and an unmet electric load of 5.2%. The other attractiveset-ups from the list that can be used for the implementation are those with 92% and 100% renewableenergy utilization, which have a net present cost of $273,549 and a cost of energy of 0.423 $/kWh for92%, and an NPC of $278,956 and a COE of 0.434 $/kWh for 100%. Even though these set-ups present anincrease in the NPC and COE from the first option, they are still cost-effective when compared againstthe cost of grid extension. For the system with 92% utilization without PV, there is an unmet electricload of 6.1%, a shortage of capacity of 9.5% and an excess electricity generation of 36.7%. Similarly, thesystem with 100% has an excess electricity production of 49.2% with an unmet electric load of 6.7%and a capacity shortage of 10%. On the other hand, the implementation of this system will help toreduce the deforestation caused by the community due to their dependence on firewood as well asminimize the health problems resulting from the smoke of the firewood, at the same time improvingthe living standard of the society of Kutur village.
Author Contributions: All authors were involved in validating and making the article error-free, as well as thetechnical outcome for the set investigation work. All authors have read and agreed to the published version ofthe manuscript.
Funding: This research activity received no external funding.
Acknowledgments: Authors would like to acknowledge the Center of Bioenergy and Green Engineering,Department of Energy Technology, Aalborg University, Esbjerg, Denmark for supporting the research activitiesand for the shared technical expertise.
Conflicts of Interest: The authors declare no conflict of interest.
Sustainability 2020, 12, 3808 13 of 14
Abbreviations
CC Cycle chargingCOE Cost of energyEEP Ethiopian Electric Power CorporationFL Fluorescent LampFTC Farmers training centerHC Health centerICS Inter-Connected SystemLCC Life cycle costLPSP Loss of power supply probabilityLV Low voltageMV Medium voltageMV Medium VoltageMW Mega-WattNMSA National Metrological Service AgencyNPC Net Present CostsPV Photo-voltaicUEAP Universal Electric Access Program
References
1. Invest in Ethiopia, Ethiopia Trade and Investment. Available online: www.ethioembassy.org.uk/trade_and_investment (accessed on 2 January 2019).
2. Drake, F.; Mulugetta, Y. Assessment of Solar and Wind Energy Resources in Ethiopia. I. Solar Energy; Universityof Leeds: Leeds, UK, 1996.
3. Dihrab, S.S.; Sopian, K. Electricity Generation of Hybrid PV/Wind Systems in Iraq; University KebangsaanMalaysia: Selangor, Malaysia, 2010.
4. Barley, C.D.; Lew, D.J.; Flowers, L.T. Sizing Wind/ PV Hybrids for Households in Inner Mongolia; Wind Power’97; National Renewable Energy Laboratory: Golden, CO, USA, 1997.
5. Fiedler, F.; Pazmino, V.; Berruezo, I.; Maison, V. PV-Wind Hybrid Systems for Swedish Locations; HögskolanDalarna: Borlänge, Sweden, 2008.
6. Mandal, S.; Das, B.K.; Hoque, N. Optimum sizing of a stand-alone hybrid energy system for rural electrificationin Bangladesh. J. Clean. Prod. 2018, 200, 12–27. [CrossRef]
7. Aberilla, J.M.; Gallego-Schmid, A.; Stamford, L.; Azapagic, A. Design and environmental sustainabilityassessment of small-scale off-grid energy systems for remote rural communities. Appl. Energy 2020, 258,114004. [CrossRef]
8. Zhao, P.; Xu, W.; Zhang, S.; Wang, J.; Dai, Y. Technical feasibility assessment of a standalone photovoltaic/
wind/adiabatic compressed air energy storage based hybrid energy supply system for rural mobile basestation. Energy Convers. Manag. 2020, 206, 112486. [CrossRef]
9. Moner-Girona, M.; Bódis, K.; Morrissey, J.; Kougias, I.; Hankins, M.; Huld, T.; Szabó, S. Decentralized ruralelectrification in Kenya: Speeding up universal energy access. Energy Sustain. Dev. 2019, 52, 128–146. [CrossRef]
10. Gabra, S.; Miles, J.; Scott, S.A. Techno-economic analysis of stand-alone wind micro-grids, compared with PVand diesel in Africa. Renew. Energy 2019, 143, 1928–1938. [CrossRef]
11. Ariyo, B.O.; Akorede, M.F.; Omeiza, I.O.A.; Amuda, S.A.Y.; Oladeji, S.A. Optimisation analysis of astand-alone hybrid energy system for the senate building, university of Ilorin, Nigeria. J. Build. Eng. 2018,19, 285–294. [CrossRef]
12. Yeshalem, M.T.; Khan, B. Design of an off-grid hybrid PV/wind power system for remote mobile base station:A case study. AIMS Energy 2017, 5, 96–112. [CrossRef]
13. Jariso, M.; Khan, B.; Tesfaye, D.; Singh, J. Modelling and Designing of Stand-Alone Photovoltaic System(Case Study: Addis Boder Health Center south west Ethiopia). In Proceedings of the ICECA 2017 IEEEInternational Conference on Electronics, Communication, and Aerospace Technology, Coimbatore, India,20–21 April 2017; pp. 168–173.
Sustainability 2020, 12, 3808 14 of 14
14. Jariso, M.; Khan, B.; Tanwar, S.; Tyagi, S.; Rishiwal, V. Hybrid Energy System for Upgrading the RuralEnvironment. In Proceedings of the 2018 IEEE Globecom Workshops (GC Wkshps), Abu Dhabi, UAE,9–13 December 2018; pp. 1–6.
15. Duffie, J.A.; Beckman, W.A. Solar Engineering of Thermal Processes, 4th ed.; Wiley: New York, NY, USA, 2013.16. Ethiopia News/GTZ Staff Magazine/August. 2010. Available online: www.enervest.de/fileadmin/downloads/
gtz060810 (accessed on 5 January 2020).17. EEPCo, UEAP. Electric Power Distribution Network Design and Construction Training Manual.18. Final Report Country Background Information Solar and Wind Energy Utilization and Project Development
Scenarios. October 2007. Available online: http://swera.unep.net/typo3conf/ext/metadata_tool/archive/
download/EthiopiaSWERAreport_258.pdf (accessed on 6 December 2019).19. Wind Power to Be Competitive with Natural Gas by 2016 Monday. 14 November 2011. Available online: http://oilprice.
com/Alternative-Energy/Wind-Power/Wind-Power-To-Be-Competitive-With-Natural-Gas-By-2016.html (accessedon 8 January 2020).
20. Renewables 2019 Global Status Report. Available online: https://www.ren21.net/gsr-2019/ (accessed on8 January 2020).
21. Hardy Diesel Generators. Available online: http://www.hardydiesel.com/diesel-generators-7-33-kw.html(accessed on 6 January 2020).
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open accessarticle distributed under the terms and conditions of the Creative Commons Attribution(CC BY) license (http://creativecommons.org/licenses/by/4.0/).